乌鸦传媒

The growing role of pediatric intensive care units (PICUs) for managing bronchiolitis in infants and young toddlers over the last decade was associated with increased use of noninvasive respiratory support, but without a corresponding reduction in invasive mechanical ventilation (IMV), a retrospective study found.

From 2013 to 2022, admissions for the common lower respiratory tract infection at 27 PICUs in the U.S. jumped threefold among children under age 2 years, from 1,706 to 5,204, reported researchers led by Jonathan Pelletier, MD, of Akron Children's Hospital in Ohio, in .

Yet while the use of high-flow nasal cannula (HFNC) as the maximum level of respiratory support increased nearly fivefold and the use of noninvasive ventilation (NIV) increased nearly sixfold, there was no reduction in the use of IMV over the 10-year study period:

• HFNC: from 668 encounters in 2013 to 3,186 encounters in 2022 (P=0.002)
• NIV: from 230 to 1,339 encounters, respectively (P=0.004)
• IMV: from 341 to 516 encounters (P=0.05)

Bronchiolitis, often the result of respiratory syncytial virus (RSV) infections, is the most of hospitalization for children under age 2, and the researchers had aimed to examine whether changes in the use of HFNC for the lower respiratory tract infections are in part to blame for the annual strain on pediatric hospital beds during the respiratory virus season.

"It seems biologically implausible that bronchiolitis severity has increased linearly for the past decade," wrote Pelletier and colleagues, making it more likely "that changes in clinical practices are at least partially responsible for the striking increase in HFNC and NIV use over the past decade."

Successful use of HFNC in the study -- meaning when children could be weaned to less-invasive respiratory support such as low-flow oxygen or room air -- increased from 79.8% in the 2013 encounters to 84% in the 2022 encounters (P=0.002). Success with NIV, meanwhile, increased from 73.2% to 84% (P<0.001).

They said that combined with prior research, the findings suggest that multiple factors are behind these increases, "including a rising population-based PICU admission rate, increased consolidation of pediatric admissions to children's hospitals, a rising proportion of children with bronchiolitis being admitted to the PICU, and an increase in the proportion of patients admitted to the PICU receiving noninvasive respiratory support."

"Thus," added Pelletier and coauthors, "solutions aimed at reducing PICU burden need to consider multiple components in the chain of care."

Prior randomized have shown lower treatment failure rates with HFNC when compared with low-flow oxygen, Pelletier and coauthors noted in their introduction. But "importantly, these trials used HFNC predominantly in acute care settings rather than exclusively in PICUs," they wrote. "In contrast, some observational studies in North America have shown that HFNC use is associated with increased PICU admissions for bronchiolitis."

Those findings, along with the fact that most previous database studies were unable to clearly separate patients who received HFNC from other forms of noninvasive respiratory support, "has made it impossible to directly examine whether increases in PICU admissions for bronchiolitis are associated with an increase in HFNC use," wrote Pelletier and colleagues.

For their study, the researchers examined 33,816 PICU encounters for children under age 2 with bronchiolitis from 2013 to 2022. The researchers relied on the (VPS) database and only included the 27 PICUs that began collecting data on HFNC use for all study years.

Overall, 59.7% of the children were boys, and 4.5% had preexisting cardiac disease. Most (94.4%) were 29 days or older while 5.6% were 28 days or younger. About two-thirds of the patients had been admitted to the PICU from the emergency department, and nearly all the children survived to discharge (99.8%). In the 15,518 cases where the pathogen was known, 49% had RSV.

Across the entire study period, 83% of the children received HFNC, 31% received NIV, 13% received IMV, and 0.2% received extracorporeal membrane oxygenation (ECMO) during their PICU stay.

Examining the highest level of respiratory support required showed statistically significant annual increases for HFNC (242 more encounters per year) and NIV (126 more per year), but not for IMV (10 more per year) or ECMO (1 more per year).

Of the 27,282 HFNC encounters over the 10-year study period, 81.8% were successfully weaned off HFNC to less-invasive respiratory support, while 1.6% were restarted on HFNC and the rest were escalated to NIV (11.2%) or either IMV or ECMO (5.4%). Median HFNC duration ranged from 20.7 to 58.5 hours across PICUs.

Of the 10,398 NIV encounters, 81.5% were successfully weaned to less-invasive respiratory support (HFNC, low-flow oxygen, or room air), while 7.6% were restarted on NIV and 10.9% were escalated to IMV or ECMO. Median NIV duration ranged from 14.1 to 71.7 hours across PICUs.

Multivariate analyses showed the following factors to be associated with a higher likelihood of HFNC or NIV failure: lower weight, cardiac disease, higher Pediatric Risk of Mortality (PRISM) III scores, and admission to the PICU from outside the emergency department.

Median PICU length of stay was 2.2 days and decreased over the study period (2.6 in 2013 to 2.0 days in 2022). The researchers also detected a small but significant decrease in PRISM III scores, findings that did not suggest that the increased use of the PICU was linked with bronchiolitis severity.

A main limitation was that the study could not extend the analysis to the approximately 200 PICUs included in the VPS database, as collecting data on HFNC use was not mandatory until 2017. Pelletier's group also noted that the use of HFNC in other hospital settings was not captured in the current analysis and that determining the severity of a low-mortality illness like bronchiolitis can be challenging in retrospective studies.

Comment